diff --git a/docs/sql-programming-guide.md b/docs/sql-programming-guide.md
index ee231a934a3a..52d6b9d73567 100644
--- a/docs/sql-programming-guide.md
+++ b/docs/sql-programming-guide.md
@@ -924,13 +924,6 @@ Configuration of Parquet can be done using the `setConf` method on `SparkSession
     flag tells Spark SQL to interpret INT96 data as a timestamp to provide compatibility with these systems.
   </td>
 </tr>
-<tr>
-  <td><code>spark.sql.parquet.cacheMetadata</code></td>
-  <td>true</td>
-  <td>
-    Turns on caching of Parquet schema metadata. Can speed up querying of static data.
-  </td>
-</tr>
 <tr>
   <td><code>spark.sql.parquet.compression.codec</code></td>
   <td>snappy</td>
@@ -1587,6 +1580,9 @@ options.
       Note that this is different from the Hive behavior.
     - As a result, `DROP TABLE` statements on those tables will not remove the data.
 
+ - `spark.sql.parquet.cacheMetadata` is no longer used.
+   See [SPARK-13664](https://issues.apache.org/jira/browse/SPARK-13664) for details.
+
 ## Upgrading From Spark SQL 1.5 to 1.6
 
  - From Spark 1.6, by default the Thrift server runs in multi-session mode. Which means each JDBC/ODBC
diff --git a/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala b/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala
index c407874381ac..f588abd463bb 100644
--- a/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala
+++ b/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala
@@ -284,11 +284,6 @@ object SQLConf {
     .booleanConf
     .createWithDefault(false)
 
-  val PARQUET_CACHE_METADATA = buildConf("spark.sql.parquet.cacheMetadata")
-    .doc("Turns on caching of Parquet schema metadata. Can speed up querying of static data.")
-    .booleanConf
-    .createWithDefault(true)
-
   val PARQUET_COMPRESSION = buildConf("spark.sql.parquet.compression.codec")
     .doc("Sets the compression codec use when writing Parquet files. Acceptable values include: " +
       "uncompressed, snappy, gzip, lzo.")
@@ -1010,8 +1005,6 @@ class SQLConf extends Serializable with Logging {
 
   def parquetCompressionCodec: String = getConf(PARQUET_COMPRESSION)
 
-  def parquetCacheMetadata: Boolean = getConf(PARQUET_CACHE_METADATA)
-
   def parquetVectorizedReaderEnabled: Boolean = getConf(PARQUET_VECTORIZED_READER_ENABLED)
 
   def columnBatchSize: Int = getConf(COLUMN_BATCH_SIZE)